Thursday, October 22, 2009
Hubert Horan, an independant aviation analyst with 25 years of experience and a previous guest writer for this blog, has submitted a rebuttal to the oneworld Alliance's recent filing against Justice Department concerns over extending antitrust immunity to global airline alliances. Horan's summary of his criticisms are bellow, followed by a link to his full filing.
The BA/AA Statement directly acknowledged the evidentiary problems highlighted by the DOJs attack on the DOTs Star Alliance/Continental approval. On the one hand BA/AA are asking for the exact same treatment as earlier ATI cases, yet they recognize that those cases were approved despite the complete absence of any legitimate, objective evidence of public benefits and minimal competitive risks. Thus the need for their Statement (OST 2008-0252-3357), because (unlike Skyteam and Star), BA/AA knew they couldnt count on DOT approval without having some sort of evidence on the case record.
However, their Statement was some of the most embarrassingly awful analysis Ive seen in my quarter-century in this business. The central arguments--that there is no statistically significant fare effect from reducing the actual number of carriers on a route from two to one and that granting antitrust immunity has no significant effect on fares are based on an incredibly simplistic regression where all of the input data is wrong. The Statement claims that ATI grants will never affect consumer pricesnot just in this narrow case, under current conditions, but anywhere, anytime, regardless of market conditions. The claims that ATI grants will never affect consumer prices is wholly based on their regression which attributes all observed price variations to variables reflecting the number of competitors. But the input data was garbage, because it treated all immunized partners (Northwest and KLM on Detroit-Amsterdam) and independent price competitors, just like BA and Continental on Houston-London. Thus none of the explanatory variables were accurate (for example, the competitors=2 variable was a random mix of markets with 1 and 2 actual competitors). Thus the regression output and the conclusions quoted above are also garbage.
The Statement separately notes a variety of competitive disadvantages that BA/AA might face absent full freedom to Collude, but while these may not be favorable to BA/AA shareholders, there is zero evidence that immunity would lead to net public benefits across the market, which is the relevant antitrust standard. Immunity may make it easier for BA/AA to harmonize frequent flyer mileage accrual rules, but that doesnt mean immunity will give BA/AA customers a more lucrative frequent flyer program. Anyone who participates in these programs knows that the claim that reduced competition leads to more generous programs is ridiculous. The Statement argues that the DOT should ignore the risks of reduced competition because there in the past immunized partners sometimes increased capacity on hub-to-hub routes, but presented no evidence that the current application would lead to overall capacity increases (and the past increases were all one-time events that occurred a decade ago under very different market conditions). Industry consolidation (especially the radical consolidation occurring on the North Atlantic) poses risks to consumers if airlines can more readily restrict supply without fear of meaningful market discipline, and the Statement asks us to ignore these risks on the basis of a totally incompetent regression. In my experience, testimony like this is usually a mix of valid, marginal and challengeable points. I couldnt find a single legitimate claim in the BA/AA Statement that had any relevance to the case at hand.
My statement attacking the BA/AA is also on the case record at OST-2008-0252-3362 (available here). Blog readers may wish to compare the two Statements and draw their own conclusions. One cannot begin to deal with the larger issues raised by the specific BA/AA application or North Atlantic consolidation more generally without first addressing the DOJs basic evidentiary problems. It is unacceptable to make major antitrust decisions based on wholly unsubstantiated assertions, or evidence totally unrelated to the case at hand, or on the basis of analysis as atrociously awful as what you'll find in the BA/AA Statement.